flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "zhuhaifeng (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-4840) Measure latency/delay of record processing and expose it as a metric
Date Thu, 20 Oct 2016 03:26:59 GMT

    [ https://issues.apache.org/jira/browse/FLINK-4840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15590636#comment-15590636

zhuhaifeng commented on FLINK-4840:

It measure the time cost of record processing from the beginning time to the completion time(eg.
from System.nanoTime() ). It's a histogram metric and a  update interface will be called to
save the processing time when a record processing completion, it's just store the processing
time. The calculate will be invloved when MetricFetcher to fetcher a snapshot. so it will
not introduce a non-trivial overhead.

> Measure latency/delay of record processing and expose it as a metric
> --------------------------------------------------------------------
>                 Key: FLINK-4840
>                 URL: https://issues.apache.org/jira/browse/FLINK-4840
>             Project: Flink
>          Issue Type: Improvement
>          Components: Metrics
>            Reporter: zhuhaifeng
>            Assignee: zhuhaifeng
>            Priority: Minor
>             Fix For: 1.2.0
> We should expose the following Metrics on the OperatorIOMetricGroup:
>  1. latency(ms): record processing time as a histogram metric.
>  2. procTimeCostPerSecond(ms): record processing time accumulated per second, it measure
record processing percentage to check if it's a main time costing
>  3. delay(ms): the time between upstream emit record and downstream start record processing

This message was sent by Atlassian JIRA

View raw message